Debian Codegraph @

doc/src/intrinsics/numeric/gamma.md

raw
1 # gamma(x): Gamma
2
3 Gamma function.
4
5 ## Declaration
6
7 ### Syntax
8
9 ```fortran
10 retval = gamma(x)
11 elemental real function gamma(x)
12 ```
13
14 ### Arguments
15
16 `x` the input value must be of type real. It should not be zero or a negative
17 integer.
18
19 ### Return values
20
21 The return value is of same type and kind as of `x`.
22
23 ## Description
24
25 **gamma(x)** computes $\gamma(x)$. For positive, integer value of `x`, the Gamma
26 function simplifies to factorial function:
27
28 $\gamma(x) = (x-1)!$
29
30 In general, if $x > 0$:
31
32 $\gamma(x) = \int_{0}^{\infty} e^{-t} dt$
33
34 and if $-n-1 < x < -n$ where n is an integer >= 0:
35
36 $\gamma(x) = \int_{0}^{\infty}(e ^{-t} - \sum\limits_{k=0}^n \frac{(-t)^k}{k!} dt)$
37
38 ## Types
39
40 Supported argument types is real.
41
42 ```fortran
43 interface gamma
44 module procedure sgamma, dgamma
45 end interface
46
47 contains
48
49 elemental real(sp) function sgamma(x)
50 real(sp), intent(in) :: x
51 end function
52
53 elemental real(dp) function dgamma(x)
54 real(dp), intent(in) :: x
55 end function
56 ```
57
58 ## Examples
59
60 ```fortran
61 program intrinsics_gamma
62 print *, gamma(0.5)
63 print *, gamma(1.0)
64 end program
65 ```
66
67 **Result:**
68
69 ```
70 1.77245
71 1.000
72 ```
73
74 ## See Also
75
76 [epsilon](epsilon.md)